Feasibility of Inconspicuous GAN-generated Adversarial Patches against Object Detection

TL;DR

This paper evaluates the feasibility of generating inconspicuous, GAN-based adversarial patches for object detection, demonstrating that pre-trained GANs can produce realistic patches without significantly degrading detection performance.

Contribution

The study adapts existing GAN-based patch generation methods to object detection, comparing training-integrated and pre-trained approaches, and assesses their realism and effectiveness.

Findings

Pre-trained GANs produce more naturalistic patches.

Using pre-trained GANs maintains detection performance.

Trade-off exists between patch realism and attack efficacy.

Abstract

Standard approaches for adversarial patch generation lead to noisy conspicuous patterns, which are easily recognizable by humans. Recent research has proposed several approaches to generate naturalistic patches using generative adversarial networks (GANs), yet only a few of them were evaluated on the object detection use case. Moreover, the state of the art mostly focuses on suppressing a single large bounding box in input by overlapping it with the patch directly. Suppressing objects near the patch is a different, more complex task. In this work, we have evaluated the existing approaches to generate inconspicuous patches. We have adapted methods, originally developed for different computer vision tasks, to the object detection use case with YOLOv3 and the COCO dataset. We have evaluated two approaches to generate naturalistic patches: by incorporating patch generation into the GAN training process and by using the pretrained GAN. For both cases, we have assessed a trade-off between performance and naturalistic patch appearance. Our experiments have shown, that using a pre-trained GAN helps to gain realistic-looking patches while preserving the performance similar to conventional adversarial patches.